|Publication number||US5309917 A|
|Application number||US 07/834,425|
|Publication date||May 10, 1994|
|Filing date||Feb 12, 1992|
|Priority date||Sep 12, 1991|
|Also published as||CA2118930A1, CA2118930C, DE69232267D1, DE69232267T2, EP0606301A1, EP0606301A4, EP0606301B1, US5443073, WO1993004627A1|
|Publication number||07834425, 834425, US 5309917 A, US 5309917A, US-A-5309917, US5309917 A, US5309917A|
|Inventors||Xiang Wang, Hun H. Sun|
|Original Assignee||Drexel University|
|Export Citation||BiBTeX, EndNote, RefMan|
|Patent Citations (45), Non-Patent Citations (28), Referenced by (234), Classifications (14), Legal Events (7)|
|External Links: USPTO, USPTO Assignment, Espacenet|
This is a continuation-in-part of U.S. patent application Ser. No. 07/758,034, filed Sep. 12, 1991, now abandoned, which is hereby incorporated by reference herein in its entirety.
The invention relates more specifically to continuous heart rate and hemodynamic monitoring.
Cardiac monitoring to which the present invention relates includes both the determination of heart rate (HR) from electrocardiogram (EKG) signals and the determination of heart stroke volume (SV) from thoracic impedance signals, from which cardiac output (CO) can be estimated.
Heart rate can be determined in a number of ways. The phonocardiogram is considered among the most accurate methods of determining heart rate. However, due to the practical difficulties in using it, the phonocardiogram method is generally not employed for any continuous or long-term monitoring.
Heart rate is most typically determined by the electrocardiogram (EKG). The analog EKG signal typically displays electrocardio events as perturbations referred to as waves. The heartbeat is most clearly reflected in the EKG signal as an R wave peak between a pair of adjoining Q and S wave valleys. The basic and commonly used method of automatically identifying the QRS wave pulses in point is the threshold method in which the rate of voltage change between consecutive data points of the EKG signal is monitored and compared with a threshold value. Slopes exceeding the threshold value are deemed to be associated with a portion of the QRS pulse. While this method commonly detects the interval between consecutive R waves successfully more than eighty percent of the time, it typically has difficulty in dealing with sources of irregular signal components such as pacemakers, muscle noise, 60 Hz interference as well as nearby T or P waves which may also be associated with significant slope changes.
Hemodynamic monitoring of the heart can provide very valuable physiological information regarding the functional state of the myocardium, which is intimately related to its mechanical behavior. The quantitative measurement of blood flow, or the cardiac output (CO), is one of the most useful parameters in assessing cardiac capability, but it is also one of the most difficult to measure. It cannot be accomplished with the electrocardiogram (EKG) which does not reflect the real pumping action of the heart.
Both invasive and non-invasive methods are available for measurement of cardiac output. The invasive methods are considered the most accurate. The risks associated with them are often an unacceptable trade-off, for they require direct access to the arterial circulation. In addition, invasive methods are not suitable for repetitive measurements and normally cannot be performed outside a hospital. Furthermore, invasive methods are very demanding in terms of time consumption and the need for skilled personnel.
Impedance cardiography has been found to be one non-invasive method with the potential for monitoring the mechanical activity of the heart with virtually no risk. It can be conveniently handled by nursing and non-technical staff. It can usually tolerate moderate patient movement and can be left unattended for continuous and long-term monitoring.
U.S. Pat. No. 3,340,867, now U.S. Pat. No. Re. 30,101, to Kubicek et al. discloses an impedance plethysmograph which employs four electrodes, two around the neck and two around the torso of a patient, to provide an impedance difference signal from the two center electrodes. The outermost pair of electrodes apply a small magnitude, high frequency alternating current to the patient while the inner pair of electrodes were used to sense voltage levels on the patient above and below the patient's heart. The impedances of the patient at each of the inner pair of electrodes could be determined from the sensed voltages and known applied currents.
According to Kubicek et al., stroke volume (SV) is related to impedance Z as follows:
where R is blood resistivity, L is the distance between the inner voltage sensing electrodes, Zo is the mean thoracic impedance determined from the inner, voltage sensing electrodes, VET is the ventricular ejection time and dz/dtmin is the maximum negative slope change of time-differentiated impedance signal, which is the time-differentiated difference between the impedances determined at the center two electrodes. The above equation is referred to as Kubicek's equation. Cardiac output per minute is stroke volume times heart rate in beats per minute.
The Kubicek equation is based upon a parallel column model of the thorax in which it is assumed:
(1) the thorax is a cylinder, consisting of two electrically conducting tissue paths, also of cylindrical form with uniform cross-sectional areas and homogenous conducting materials, the first path being the blood with a relatively low resistivity and the second path being all other tissues with relatively high resistivities;
(2) the relationship between the maximum impedance change and the stroke volume during the cardiac cycle is linear;
(3) impedance measurements of the individual specific tissue volumes are not very useful in developing the model (the parallel columns model relies on the intact thoracic measurements); and
(4) at 100 kHz frequency, a physiologically safe frequency, the relative thoracic impedance changes are at a maximum, the effects of polarization are negligible, and the reactive component of impedance is small, especially when compared to the real component, so that the reactance could be ignored in determining impedance with only a small error.
Yet another model and equation for relating impedance and stroke volume has been proposed by Sramek. According to Sramek, stroke volume (SV) is related to impedance Z as follows:
SV=[(O.17H)3 /4.2]*[VET]*[dZ/dtmin /Zo ]
where H is the patient's height. The Sramek equation is based upon a frustoconical model of the thorax. The Sramek model illustrates some improvement and accuracy over the Kubicek model but its major assumptions are still similar to those of the Kubicek model.
Despite its advantages, impedance cardiography has not been well accepted by clinicians for three primary reasons:
(1) poor correlation of the methods and models with the results of the more accepted invasive techniques;
(2) still a relatively high dependance on skilled technical operators; and
(3) still a discomfort to and/or disturbance of patients associated with the use and application of band electrodes.
It is believed that poor correlation, the primary reason, can be traced back to a single source, namely the continuing inability to relate impedance cardiography and its mathematical model directly to cardiac physiology.
The following are definitions and abbreviations of some of the terms used frequently herein:
Heart Rate (HR): the number of times the heart contracts each minute.
Ventricular Ejection Time (VET): the time interval of the opening and closing of aortic value during which there is movement of blood out of a ventricle.
Stroke Volume (SV): the volume of blood pumped out by a ventricle (in particular the left ventricle) with each contraction of the heart.
Cardiac Output (Co): the amount of blood pumped out of the heart into the systemic circulation each minute.
Ejection Fraction (EF): the ratio SV/EDV, which is the percentage of blood in a ventricle ejected with each contraction; it is directly related to the strength of the heart with <50% considered abnormal.
End Diastolic Volume (EDV): the volume of blood that fills the ventricle before ejection.
It would be desirable to determine heart rate more accurately than can be determined using the cardiogram threshold method currently employed.
It further would be desirable to provide noninvasive, cardiographic impedance monitoring to estimate stroke volumes, cardiac outputs and related cardiac function parameters which correlate more closely with the stroke volumes, cardiac outputs and the like determined by means of recognized, accepted invasive procedures, but which does not require of operators the technical skills required by current impedance cardiograph systems, and does minimize discomfort to the patient on which the system is used, thereby permitting relatively long-term monitoring of the patient's condition.
One aspect of the invention is a method of processing a time-derivative, thoracic impedance signal generated from a patient to identify events in the time-derivative signal associated with beats of the patient's heart. The method comprises the steps of generating a time-frequency distribution of data from the time-derivative impedance signal for a selected portion of the time-derivative impedance signal spanning a single heartbeat of the patient and identifying a time of occurrence of at least one event in the selected portion of the time-derivative signal from the time-frequency distribution.
Another aspect of the invention is a method of processing an electrocardiogram signal data from a patient. The method comprises the steps of differentiating the electrocardiogram signal data with respect to time, scaling the differentiated signal nonlinearly to emphasize amplitude peaks in the differentiated signal data; identifying a maximum amplitude value of the scaled signal data derived from an interval of the electrocardiogram signal selected to include several consecutive seconds of data for several consecutive heartbeats of the patient; identifying each amplitude value of the scaled signal data in the selected interval at least as great as a predetermined fraction of the maximum amplitude value; and generating a coefficient of variation for the amplitude values of the scaled signal identified from the selected interval in the previous step.
Another aspect of the invention is a method of determining stroke volume of a patient's heart from thoracic impedance measurements. The method comprises the steps of applying a fluctuating excitation current to a patient through a first pair of electrodes located on opposing upper and lower sides of the patient's heart; generating from a second pair of electrodes located on opposing sides of the patient's heart between the first pair of electrodes a first signal related to impedance of the patient at the first electrode of the second pair and a second signal related to impedance of the patient at the second electrode of the second pair; identifying an impedance from each of the first and second signals occurring simultaneously with one heartbeat of the patient; and multiplying together the identified impedances of each of the first and second signals to estimate a cardiac parameter at least related to stroke volume of the patient's heart during the one heartbeat.
In the drawings, like numerals are employed to indicate like elements throughout.
The foregoing summary as well as the following detailed description of the preferred embodiments of the invention will be better understood when read in conjunction with the appended drawings. For the purpose of illustrating the invention, there is shown on the drawings embodiments which are presently preferred. It should be understood, however, that the invention is not limited to the precise arrangements and instrumentality shown.
FIG. 1 depicts the cardiographic monitoring system of the present invention with the electrodes thereof applied to a patient P;
FIG. 2a depicts a portion of an exemplary EKG signal spanning a single heartbeat;
FIG. 2b depicts a portion of an exemplary inverted dz/dt signal spanning a single heartbeat;
FIG. 2c depicts a portion of an exemplary phonocardiogram signal spanning the same heartbeat shown in the dz/dt signal;
FIGS. 3a-3b depict in block diagram form, the steps followed by the processor of the system in determining the various cardiac parameters identified and calculated by the system;
FIG. 4 depicts schematically a linear phase delay filter for combined low-pass, high-pass filtering;
FIG. 5 depicts in block diagram form the detailed steps of the EKG signal processing;
FIG. 6 depicts in block diagram form the detailed steps of the time differential thoracic impedance signal processing; and
FIG. 7 depicts a two-dimensional representation of local frequency change determined from an exemplary Spectrogram time-frequency distribution derived from the time differential impedance signal.
In the various figures like reference numerals are used to refer to like elements. There is shown diagrammatically in FIG. 1 a preferred cardiographic impedance monitoring system according to the present invention, which is indicated generally at 20 and is coupled to a patient P for use. System 20 preferably includes a first, outer pair of electrodes 21 and 24, 21 being a strip electrode and 24 being a band electrode, and two pairs of parallel connected, spot-type skin electrodes, indicated generally at 22 and 23. The patient's heart is indicated diagrammatically, in phantom, at H. The first pair of electrodes 21 and 24 are applied to the patient's skin on opposite upper and lower sides of the heart H, preferably equally spaced from the heart. The pairs of parallel coupled spot electrodes 22 and 23 are applied to the patient on opposite upper and lower sides of the heart H respectively and between the first or outer pair of electrodes 21, 24. Each pair of electrodes 22 and 23 is preferably positioned on opposite lateral sides of the patient at uniform heights above and below the heart H. Again, electrode pairs 22, 23 are preferably equally spaced from or otherwise symmetrically positioned with respect to the heart H. A pair of conventional, electrocardiogram electrodes 25a and 25b are further provided.
Preferably, all five electrodes 21-25 are coupled through a signal pickup and preprocessor 26 and then through an analog to digital converter 28 to a data processor 30. Preferably a color video monitor 32 and an optional hard copy printer 34 are provided under the control of the processor 30. While a separate pair of electrocardiogram electrodes 25a and 25b is shown, it will be appreciated that the electrode pairs 22 and 23 might also be configured to generate an electrocardiogram signal by suitable processing of their signals. The outer pair of electrodes 21 and 24 are used to apply a fluctuating, preferably alternating current through the patient P between those electrodes 21 and 24 while electrode pairs 22 and 23 are provided to sense voltage levels on the patient P from which thoracic impedance is determined, as will be discussed.
The general advantages of tetrapolar electrode configurations, such as that of system 20, have over bipolar electrode configurations are that the voltage sensing electrodes 22 and 23 are substantially free of skin impedance, can measure impedance with less electrode-interface artifact and offer the possibility of providing a more uniform current density distribution in the segment of interest in the patient.
There is a distortion in the current density distribution in patients in the vicinity of the current electrodes 21 and 24, known as the edge effect. In addition to edge effects, uniform current distribution is disturbed by the superposition of a relatively highly conductive sphere, namely the heart, and the relatively low conductivity of air in the lungs in the region between electrodes 22 and 23. The current distribution has been found to be more uniform in the central region between the electrodes 21 and 24 the more widely spaced those electrodes are. In addition, it has been found that the positions of the current electrodes 21 and 24 relative to those of the voltage detecting electrodes 22 and 23 are further important to ensure uniform current distribution. It has been found that if the distances between adjoining pairs of the electrodes 21 and 22, and 23 and 24 are sufficiently great, at least about 3 cm and, preferably, about 4 cm or more, the edge effect is essentially minimized and the impedance measurement becomes stable.
The upper current electrode 21 is preferably applied to the patient's forehead while the upper voltage sensing current and pick-up electrodes 22 are applied to the patient's neck. This configuration has several advantages over previous systems employing neck current and pickup electrodes. It is a simple configuration for the technician to remember while assuring that the upper two electrodes 21 and 22 are sufficiently separate to avoid edge effects. The further spacing of the upper current electrode 21 from the heart provides a more uniformly parallel current distribution and equal potential lines through the patient, particularly in the area between the measuring electrodes 22 and 23. In addition, this arrangement minimizes the problem of attempting to locate all upper electrodes 21 and 22 on the patient's neck where catheters and bandages are often encountered. The forehead is usually untouched by medical devices. All four electrodes 21-24 are preferably mounted to the patient in parallel planes to optimize the uniformity of the current distribution and the location of the sensing electrodes 22 and 23 along the equal potential lines.
Band electrodes have been found to provide the most parallel current lines and equal potential distributions through the patient and therefore should, at least theoretically, give the best results. There are several disadvantages to band electrodes. They may be difficult or impossible to apply to patients with extensive chest wounds because of the presence of dressings, tubes, etc. They require relatively long polarization time to stabilize before measurements can be taken and, therefore, may not be ussable in emergency situations. They are more sensitive to motion artifact due to the large area of contact of the electrode. Lastly, they are generally more uncomfortable for patients when used for long-term monitoring. The advantages and disadvantages of spot electrodes are just the reverse of those of the band electrodes.
The correlation between the results of using an all band electrode configuration and the mixed band and spot electrode configuration shown in FIG. 1 were distinctly higher (correlation coefficient equal to about 0.95) than the correlation between an all band and an all spot electrode configurations (correlation coefficient equal to about 0.75). The combined band and spot electrode configuration shown in FIG. 1 appears to have all the advantages of each type with almost no disadvantages of either. In particular, the mixed electrode system supplies parallel, iso-current lines and parallel equal potential lines generally perpendicular to the iso-current lines. The system is easy to apply to patients, even with extensive chest wounds, dressing, tubes or the like on the chest. It has relatively short polarization time and thus can be used in emergency situations. It is less sensitive to motion artifact due to the small area of contact provided by spot electrode pairs 22 and 23 which provide the impedance signal. The spot electrodes 22 and 23 are also more comfortable to the patient.
The signal pick-up and preprocessor 26 is preferably a Minnesota Impedance Cardiograph (MIC) Model 304B, supplied by Sorocom, Inc., of Minnesota. The MIC 304B provides a high frequency (approximately 100 kHz), low amplitude (4mA RMS) alternating current at pick-ups provided for the electrodes 21 and 24. Pick-ups are also provided for the parallel coupled voltage sensing electrode pairs (22) and (23) and for the EKG electrodes 25a, 25b.
The preferred MIC preprocessor generates and outputs four analog signals: the mean thoracic impedance signal (Zo), the change in thoracic impedance signal (delta Z or ΔZ), the time-derivative impedance signal (dZ/dt) and the electrocardiogram signal (EKG). The mean thoracic impedance signal, Zo, is the impedance difference sensed between electrodes 22 and 23. The change in thoracic impedance signal, delta Z, is an amplification of the original Zo signal from which the DC component has been removed. The time-derivative impedance signal, dz/dt, is the time derivative of the amplified delta Z signal. The EKG signal is conventional. The four signals are shown diagrammatically on monitor 32 in FIG. 1 while expanded versions of the dz/dt and EKG signals are shown in FIGS. 2A and 2B, respectively.
FIG. 2a depicts a portion of an exemplary EKG signal while FIG. 2b depicts a portion of an exemplary inverted dz/dt signal. FIG. 2c depicts a portion of an exemplary phonocardiogram signal corresponding to the dZ/dt signal. The dZ/dt signal is conventionally inverted so that its maximum slopes will appear positive, thereby enabling the clinician to observe the cardiac event in a more familiar manner.
Cardiac events appear in the impedance and EKG signals as perturbations or "waves". EKG waves related to each heartbeat are also noted on the EKG signal of FIG. 2a at P, Q, R, S and T. The waves which are related to a single heartbeat and which appear in the dZ/dt signal are indicated in FIG. 2b at A, B, C, F, X and O. Heart sounds (S1 and S2) associated with the compression and relaxation, respectively, of the heart during a single beat, are overlaid in the dZ/dt signal of FIG. 2B. In many cases, however, the identification of individual waves in either signal is not so apparent.
The four analog signals from the preprocessor 26 are passed to the analog digital converter 28. Preferably, the A/D converter 28 is configured for differential conversion at a sampling rate of 500 Hz for each of the four analog signals with a twelve bit resolution in offset binary format. The analog to digital converter may be, for example, a Data Translation Model DT 2811/PGH.
The four digitized signals are preferably passed into an allocated memory indicated diagrammatically at such as a hard disk or RAMDISK in or associated with the processor 30 and stored in binary format for subsequent processing. The processor 30 preferably also formats the binary data signals for real time or essentially real time display on the video monitor 32. At least the dZ/dt and the EKG signals from preprocessor 26 are displayed for the clinician on the monitor 32 but the Zo signal and the delta Z signal are also preferably simultaneously displayed for the clinician who can check the validity of the various signals as they are being acquired. Preferably, system 20 processes the impedance and EKG signal data to determine at least the heart rate (HR) and the cardiac output (CO) of the patient in at least near real time and preferably displays those values as well. These and other values determined by the process may be printed out on the hard printer 34.
FIGS. 3a-3b depicts in block diagram form, the operational steps of the processor 30 in determining heart rate, cardiac output and optionally a variety of other parameters reflecting cardiac performance begins the generation of data. Preferably, the system 20 is configured to be operable in a predata signal processing mode in which the preprocessor 26 and ADC 28 operate to pass binary data signals to the processor 30 which formats the signals for real-time or essentially real-time display on monitor 32. After the clinician has verified successful signal gathering and digitization through the display, the data signal processing for cardiac parameter determinations are begun. Preferably the processor 30 will initialize the necessary calculation variables and begin storing the binary EKG and impedance signal data in a hard disk or RAMDISK in the binary format for subsequent processing.
Initial data processing includes converting the stored binary impedance and EKG signal data into decimal integer format and reading the decimal format data into a working memory area of the processor 30 in predetermined time period blocks for processing. The EKG signal data is first processed by low-pass (LP) and high-pass (HP) filtering, then differentiated and thereafter non-linearly scaled or transformed. The filtered, differentiated, scaled EKG data is then further adaptively processed to detect the R wave peaks, which are used to identify the cardiac cycles of the heartbeat, and to validate the detected peaks. If valid, the processor 30 stores the time of occurrence of each validated peak as the peak of an R wave in the EKG signal block.
In particular, preferably each consecutive five second block of the processed EKG signal data is searched for its maximum peak. The maximum peak is multiplied by a predetermined fractional value, preferably 0.5, to generate a tentative threshold. The block of data is searched to identify peaks that are greater than or equal to this threshold, with each consecutive pair of peaks being separated by a predetermined time interval, preferably in the range of 0.28 to 4.5 seconds. The coefficient of variation ("C.V."=Standard Deviation/Mean) of the peak time intervals is determined and compared to a predetermined ratio, preferably 0.3. If the determined C.V. is less than or equal to this predetermined ratio (0.3), the identified peaks are accepted as the R wave peaks. If the determined C.V. is larger than this ratio (0.3), and if the number of intervals that are less than the mean is more than the number of intervals that are greater than the mean, then the tentative threshold is reset by multiplying the maximum EKG value by a smaller predetermined fractional value, preferably 0.4. If the reverse is true, i.e., if the determined C.V. is larger than the predetermined ratio (0.3) and the number of intervals that are less than the mean is less than the number of intervals greater than the mean, then the tentative threshold is reset by multiplying the largest EKG value by a larger predetermined fractional value, preferably 0.6. The process is repeated to determine the C.V. of the newly identified intervals, as was done with the originally tentative threshold. If the determined C.V. still exceeds 0.3, the data block is discarded and the next block of processed EKG signal data is examined.
After checking for control flags that may have been generated during EKG signal block processing for the prescribed C.V. value, the processor 30 processes the data from the impedance signals, particularly the dZ/dt signal. Briefly, the R wave peaks identified from the EKG signal are used to "window" or define blocks of the dZ/dt data spanning a single heartbeat for processing. A portion of the time-derivative impedance signal dZ/dt is selected to span one heartbeat of the patient. The time-derivative impedance signal preferably is smoothed by low-pass and high-pass filtering and a time-frequency distribution of the filtered data from the windowed portion of the time-derivative impedance signal is generated in a manner to be described. Preferably, the time-frequency distribution is thereafter analyzed, preferably in conjunction with the time-derivative impedance signal, to identify from the time-frequency distribution a time or times of occurrence of at least one, and preferably two, cardiac events in the selected portion of the time-derivative signal. If the processor 30 is unable to identify the waves being sought, an artifact warning control signal is generated to set a control flag. Otherwise, the pertinent parameters from the various impedance signals, as well as related values derived from those signals including, but not limited to, the mean impedance Zo at the peak of the heartbeat contraction, dZ/dtmin, heart rate (HR), ventricular ejection time (VET), and all other parameters which may be identified or determined are moved into memory for storage. The process is repeated for each heartbeat identified in the interval of the processed EKG signal, which was originally selected to include the several consecutive seconds of data with several consecutive heartbeats of the patient. When the various parameters have been determined for each heartbeat of the interval (QRS Empty), the processor sets control flags for any errors noted and performs any remaining calculations. The processor then converts the identified and determined values into floating point numbers. Lastly, the processor preferably displays some or all of the identified and determined values on the monitor 30 and prints them on the hard copier 32, if desired.
A distinct advantage possessed by the present system over prior known systems is the filtering method employed for processing the digitized EKG and impedance signals.
Preferably, all filtering is digitally performed by processor 30 and the filters used are designed to employ only integer coefficients. This allows the filters to operate in near real-time on a relatively simple microprocessor such as an IBM® or IBM® compatible personal computer, which is also employed to perform the remaining processing steps.
The all-integer filter designs of the present invention are both simple to program and fast to execute. They have proven to be more than capable of handling at relatively high speed the data filtering required by the present system. They utilize a high-level, C language without the need for assembly language programming or a co-processor. The filters of the invention use only a relatively small number of multiplier and additional components, all with integer coefficients. A brief derivation of the filters follows.
The usual auto-regressive moving average (ARMA) system can be represented by the equation:
y[n]=a1 y[n-1]+. . . +am y[n-m]+bo x[n]+. . . +bk x[n-k]
For a low-pass filter, consider a special case of the moving average (MA) system:
Its transfer function is:
where z=esT and T is the sample interval. The k zeros in the z-plane are the roots of:
Consider the case k=12 for the z plane represented by a real abscissa (X axis) and an imaginary ordinate (Y axis), i.e. Z(x,y)=Z(real, imaginary). If a zero at z=(1,0) of the unit circle for the real and imaginary values in the z-plane, respectively, is cancelled, then the following low-pass filter transfer function is obtained:
This transfer function yields the recursive relation:
To improve the side lobes (-14 dB), second or third order zeros and poles may be taken:
H(z)=(1-z-k)2 /(1-z-1)2, or
which greatly improves the sidelobes (-27 dB, -42 dB). The recursive equation becomes:
The total order of this system has become 2k (or 3k). For k=12, the total system needs to preserve only 24 (or 36) data points.
By moving the cancelling pole z=(1,0) to z=(-1,0), a high-pass filter is obtained with the following transfer function:
and the following recursive equation:
Cancelling the zeros on the unit circle at any angle q in the z-plane other than z=(1,0) and z=(-1,0) will give a band pass filter. However, if only integer coefficients are desired, some restrictions should apply since any pair of those zeros will result in the cosine function: ##EQU1## Accordingly, angle q must be 60°, 90° and 120°, correspondingly located at 1/6T, 1/4T and 1/3T. The bandwidth can then be managed by the integer k and T.
An example of a band pass filter with a center frequency f=1/4T and nominal bandwidth NB=1/6T, would include the following transfer function and recursive equation:
H(z)=(1-z-12)2 /(1+z-2)2 and
Consider an ideal, continuous time differentiator and its respective frequency response: ##EQU2## Since the input signal is restricted to be bandlimited, it would be satisfactory if the continuous response could be: ##EQU3## The corresponding discrete-time differentiator has the following frequency response and is periodic with period 2π:
It can be shown that the corresponding impulse response of this discrete frequency response can be represented as:
which is zero for n=0 and as follows for n not equal to zero):
For example, a 6-point differentiator would be represented as:
A preferred, low-pass auto-regressive moving average filter is employed having a nominal bandwidth=±33.3 Hz, Sidelobes=-27 dB, K=15, T=500 Hz/sec, Delay=15T, Gain=225, having the following transfer function and recursive formula:
H(z)=(1-2z-15 +z-30)/(1-2z-1 +z-2)
A preferred high-pass ARMA filter is employed in the system having a bandpass=±5 Hz, and sidelobes=-14 dB. Because of the T restriction for the specific high-pass filter, a low-pass filter is designed first and then subtracted by means of a linear phase delay filter, as shown in FIG. 4. The low-pass filter is represented by the following transfer function and recursive formula, respectively:
where K=100, T=500 Hz/sec, Delay=100T, and Gain=100.
The preferred delay filter used in FIG. 4 is represented by the recursive formula:
where d[n] represents the high-pass filter output samples.
A preferred moving average (MA) differentiator used by the system 20 for differentiating the EKG signal data after low-pass/high-pass filtering has a linear slope bandwidth =±35 Hz and the following transfer function and recursive formula:
H(z)=-z-4 -2z-2 +2z+2 +z+4
where K=4, T=500 Hz/sec, and Delay=4T.
FIG. 5 depicts diagramatically in block diagram form and greater detail, the sequence of steps preferably followed by the processor 30 to process the digitized EKG data and to identify the R wave pulses. The EKG data stored in binary form is converted into digital data in blocks containing the continuous EKG signal data for a predetermined period of time. The blocks are selected to include several consecutive seconds of data from the EKG signal for several consecutive heartbeats of the patient. Five second long blocks of EKG signal data are presently preferred as such blocks are sufficiently long to permit the elimination by filtering of artifacts which might be caused by arrhythmia or patient movement and yet are sufficiently short that they still permit the various digital filters of the system to handle heart rates of between about 13 and 214 beats per minute.
Processor 30 preferably first converts the block of binary data into digital format values. It then low-pass filters the decimal format data, high-pass filters the low-pass filtered data, and applies the twice-filtered data to the previously described differentiator or derivative filter. The filtered, differentiated data is then non-linearly scaled, preferably by squaring for convenience, although raising the data by any power greater than one will provide non-linear scaling. The preferred low-pass ARMA filter has a nominal bandwidth equal to ±33.3 Hz to remove the components of the EKG signal which may be provided by external electrical interference, specifically 60 Hz AC line and other higher frequency interferences. The preferred high-pass ARMA filter has a band pass equal to ±5 Hz to eliminate unwanted low-frequency components representing slow motion artifacts such as respiration and other physical patient movements and such anomalies as arrhythmia. The preferred MA differentiator has a linear low bandwidth equal to ±35 Hz to extract and emphasize data concerning the slope change of the EKG signal. The non-linear transformation is preferably significantly great to provide an R/P equal to -75 dB where R and P stand for the amplitudes of the R and P waves, respectively, of the EKG signal associated with each heartbeat. Squaring has proved adequate. The filtered, differentiated, non-linearly scaled five second block of EKG signal data will include a number of peak values, the greatest of which will correspond to the separate R wave peaks of the original EKG signal.
According to an important aspect of the present invention, an adaptive threshold is set for each multisecond block of EKG data. This is done by identifying the magnitude of the maximum amplitude value of the filtered, differentiated, non-linearly scaled data of the block, setting a predetermined fraction of that magnitude, for example, 0.5, as a tentative threshold. All data points in the block of filtered, differentiated, non-linearly scaled data having an amplitude at least as great and preferably greater than that predetermined fractional threshold are identified and processed to determine the average of the identified amplitudes and a standard deviation of the identified amplitudes. The processor thereafter determines the coefficient of variation (CV) of the identified amplitudes, where CV=standard deviation/mean. If the coefficient of variation exceeds a predetermined value, preferably 0.3, none of the identified amplitudes are accepted or validated and that block of EKG data is preferably discarded. If the coefficient of variation of the amplitudes is equal to or less than the predetermined fraction, the EKG data is further processed.
A bi-level timing signal preferably is generated from the times of occurrence of those identified amplitudes remaining after the foregoing processing and is output as a pulsed, heart rate signal, which can be displayed on the monitor 30 and/or recorded.
Testing of this process on a large number of patients, including some with pacemakers and some exercising on a stationary bicycle, was found to provide a correct R wave detection rate for more than ninety-five percent of the EKG signals where a coefficient of variation of 0.3 was employed.
Next, the processor 30 preferably processes a block of the impedance signal data overlapping the selected block of processed EKG data to identify cardiac events reflected in the impedance signals from which the Ventricular Ejection Times (VET) may be estimated.
The VET plays an important role in impedance cardiography since it represents the time period between the opening and closing of the aortic valves during the systole-diastole cycle of the heart. It has not been clear how the VET should be defined with respect to the time-derivative impedance signal dz/dt. Several alternative definitions have been proposed including:
(1) the distance between two zero crossing points of the dz/dt signal;
(2) the distance from a first zero crossing of the dz/dt signal, before the occurrence of the "X" wave, to the "X" wave of the dz/dt signal;
(3) the distance between the 0.15* dz/dtmin point and the X wave of the dz/dt signal;
(4) the distance between the A and X waves of the dz/dt signal; and
(5) the distance between the B and X waves of the dz/dt signal.
The latter two definitions are baseline independent. The last of these definitions is presently preferred because the B and X waves are considered to represent the onset points for opening and closing of the valves and thus, physiologically correspond most closely to the end points of the theoretical interval being determined. However, this last definition is perhaps the definition least used because it is the most difficult one to detect. The B and X points often, if not usually, disappear into the noise of the basic dz/dt signal.
According to another important aspect of the invention, a time-frequency distribution analysis is generated to extract detailed information on the transient behavior of the non-stationary dz/dt signal. This leads to the concept of a mixed time-frequency signal representation.
A general class of time-frequency distributions was introduced by Cohen and have the following form: ##EQU4## where f(u) is the time signal, f*(u) is its complex conjugate, and φ is a kernel function which represents the particular distribution function selected. Several distributions have been derived from this Cohen distribution including: the Rihaczek distribution, the Page distribution, the Levin distribution, the Wigner distribution, the cumulative decay or attack spectrum and the Spectrogram distribution.
Preferably, a Spectrogram time-frequency distribution of the digitized dZ/dt signal is used by the processor 30. The Spectrogram distributors, sometimes called the time-dependent Fast Fourier Transform (FFT), is defined generally as: ##EQU5## where f[n] is the digitized dZ/dt time signal data and w[n] is a window function. Window functions particularly useful with Fourier transforms include rectangular, Bartlett (triangular), Hanning, Hamming, and Blackman. Preferably, a Hamming window function is used having the following equation: ##EQU6## Variable n=0,1, . . . M-1 and corresponds to data points of the digitized dZ/dt signal; M is the total number of data points (n) in the interval Ri to Ri+1.
When multiplied by a window function, the one-dimensional discrete signal f[n] (i.e. the digitized dZ/dt signal) is converted into a two-dimensional function of time variable and a frequency variable. Its time-dependent Fourier transform can also be interpreted as the Fourier transform of f[n+m] as viewed through the window w[m]. The window has a stationary origin and as n changes, the signal slides past the window so that, at each value of n, a different portion of the signal is viewed.
A major advantage of time-frequency analysis utilizing a Spectrogram distribution is that for the purposes of the present system 20, such an analysis is possible with the computational power of the preferred microprocessor 30. The corresponding disadvantage of a Spectrogram analysis is that there is a trade off between time and frequency resolution. An underlying assumption in using a window is that the spectral characteristics of the signal being viewed can be considered reasonably stationary over the window's duration. The faster the spectral characteristics of the signal change, the shorter the window duration should be. A shorter window length provides a higher resolution in time changes. However, the shorter the length of the window duration, the lower the frequency resolution will be. This is because the resolution of narrowband components will be sacrificed as window duration decreases. To increase frequency resolution, the window length must be increased.
The primary frequency components of the power spectrum density of typical thoracic impedance signals are found scattered approximately within the range of 2-40 Hz. However, the present system and method is preferably interested only in the contribution within the range of approximately 30-55 Hz, corresponding most closely to the frequency of opening and closing of the aortic valve.
The Spectrogram function S is preferably calculated for each R--R interval. To be certain that the Spectrogram function captures the pertinent frequency data in the dZ/dt signal associated with each heartbeat, Fast Fourier Transforms used to generate the Spectrogram are preferably determined for each heartbeat beginning at a point in time before the occurrence of the heartbeat. Preferably, the time period over which the Spectrogram function is calculataed begins before the R wave peak Ri of the heartbeat in question, preferably about twenty percent of the time interval between the heartbeat in question (Ri) and the next heartbeat(R1+1). For simplicity, reference will hereafter be made to R--R intervals but it will be understood where it is not stated that these intervals are shifted slightly backwards in time for the Spectrogram function computations. Preferably, the heartbeat times and intervals (R--R intervals) are defined by the previously derived R peaks and/or heart rate signal. The FFT is preferably calculated with the low-pass/high-pass filtered (smoothed) dZ/dt(n) data.
The following variable names have been employed in FIG. 6:
______________________________________Ri : The time location of a particular R spike of the EKG signal.Ri+1 : The time location of another R spike immediately after Ri of the same EKG signal.M: The number of dZ/dt sample points between the Ri and Ri+1 locations (M = Ri+1 - Ri) and is a varying number.D: The number of points the Ri and Ri+1 locations are shifted backwards in time.B: Location of B wave peak of the dZ/dt signal in the Ri to Ri+1 interval.X: Location of X wave peak of the dZ/dt signal in the Ri to Ri+1 interval.P: Location of dZ/dtmax within the Ri to Ri+1 interval.w[n]: A window function of data points "n".H[n]: Fourier transform value for a frequency value that corresponds to n.L: The length of an FFT process.m: The number of points or STEP, that a 2*L duration moves ahead (to the right) in the time mdomain within the Ri to Ri+1 cycle. Note that m<<2*L and preferably equals 1.S: Spectrogram function.______________________________________
The signals are processed R--R cycle by R--R cycle. Again, each R--R cycle is purposely shifted backwards early in time by D points. D is preferably twenty percent of the total number of Ri+1 -Ri data or sample points. This is to assure that the A and B points of the heartbeat associated with Ri are located inside the process cycle. The preferred sample rate is 500 Hz/sec and per channel.
Physiologically, within every process cycle Ri to Ri+1, there is a peak dZ/dt value. This peak value is dZ/dtmin when referring to the time derivative of the impedance signal and dZ/dtmax when referring to a time derivative of an inverted impedance signal. In the context of the present invention, these subscripts merely identify the sign of the differentiated impedance signal. It is assumed that the dZ/dtmax is always the positively largest one within the cycle, which is true physiologically. This dZ/dtmax can be easily detected from the filtered (inverted) dZ/dt signal data due to the fact that it has the largest positive magnitude. The location of dZ/dtmax within the cycle is detected and stored as P. Variables "i" and "k" are initialized. Variable "i" is the location pointer of the first point of each 2*L period of data copied for processing while variable "k" is the kth value of the copied 2*L data period. The first block of the flow chart of FIG. 6 indicates that Ri, Ri+1, M and P values are available.
In the second block, the first 2*L dZ/dt data points (starting from Ri -D) are copied into a memory buffer. Note that L must be an order of two and is preferably at least sixteen. In other words, it can be sixteen, thirty-two, sixty-four, one-hundred-twenty-eight, etc. However, the relationship 2*L≦M must hold. Preferably, L equals thirty-two.
In the third block, each of these 2*L dZ/dt data points are multiplied by a selected window function (w[n]), preferably a Hamming window function, also with a length 2*L. The purpose is to smooth the dZ/dt data to reduce the edge effect. A Rectangular window function might alternatively be used. These two window functions are described below:
w[n]=1 for i≦n≦i+2*L,
w[n]=0.54-0.46cos(2*π*n/M) for i≦n≦i+2*L,
In the fourth block, after the impedance data has been multiplied with the selected window function, a Fast Fourier Transform (FFT) is performed over these 2*L points. Note that the FFT routine requires 2*L points to return L points because the result has both real and imaginary parts and each part has L points.
Generally, the mathematical equation of the discrete Fourier transform for a frequency value that corresponds to n can be described as follows: ##EQU7## where: h[p] is the input function, N is the number of input points (i.e., N=2*L), p is the discrete variable in time domain (expanding from 0 to N-1), n is the discrete variable in the frequency domain and equals (0,2*L-1)) or (-N/2,N/2) or (-L,L), all are equivalent.
The result of a complete Fourier transform is a complex array:
H[-2*L/2], . . . , H[-1],H[09 ,H, . . . , H[2*L/2].
where 2*L/2 corresponds to the one-half sample frequency and 0 corresponds to zero frequency. Since the negative frequency part of the transform is only a mirror image of the positive part, it is usually dropped from the computation.
Here the input function is: h[p]=w[p]*dZ/dt[p].
For a Hamming window function, the discrete Fourier transform is given by: ##EQU8## where n=(-N/2, N/2) or (-L,L). Note that the limit of p normally might be Ri ≦p≦Ri +2L-1. However, since every 2*L duration is processed individually, as far as the Fourier transform is concerned, p always begins at 0.
The actual calculation is done by using the FFT computation algorithm to gain faster speed and the negative frequency part is not returned in the result. However, the principle and outcome are the same. Note that the values of H[n] are complex in format, i.e., real and imaginary. For -L<n<L, the set H generates two terms for every frequency, one for the real part, another for the imaginary part. Thus the actual result of this computation is an array of integers in the format:
H,H,H,H,H, . . . H[2*L-2],H[2*L-1].
where H and H are the real and imaginary part of the 0 frequency, H and H are the real and imaginary part of the next immediately higher frequency, H and H for the subsequent immediately higher frequency, and so on.
In our application, only the magnitude of the FFT is used to generate the Spectrogram function S. That is also determined in the fourth block according to the equation: ##EQU9## and n=0,1, . . . L-1.
Note that each of the magnitude points "n" corresponds to a particular frequency and thus reflects the distribution of that frequency in this 2*L portion of the signal. The bigger the magnitude is, the larger that frequency contributes. In this case, 0-15 points are generated for each Spectrogram function point k and correspond to 0-fo/2 where fo is the sample frequency. Since we chose 2L=32, 0-31 represents 0-fo, and 0-15 represents 0-fo/2. The part fo/2-fo is just the mirror image of 0-fo/2.
In the fifth block, an individual point S[k]of the Spectrogram function is determined. A certain range of frequencies contribute to the B, C and X points on the dZ/dt signal, and that frequency range is known to be between about 30-55 Hz, which roughly corresponds to frequency points between n=2 and n=4 (about 31.25 and 46.875 Hz). This range is described in the figure as frequency range W1-W2. Similarly, points n=0 to n=2 are described in the figure as frequency range W0-W1 and the points n=4 to n=15 as frequency range W2-W3.
When 0≦i≦P (P being the location of dZ/dtmax), the magnitudes of points between W1-W2 are summed together, and are normalized by the sum of the magnitude of points between W2-W3. When P<i≦M+L, the magnitudes of points between W1-W2 are summed together and are normalized by the sum of the magnitude of points between W0-W2. S[k] contributes only one point to the time-frequency distribution at time location k.
In the sixth block, k is increased by 1. However, i may be increased by a step size m to the right in the time domain. Note that m<<2*L. Thus, the second 2*L duration of the signal to be processed actually overlays with the first duration.
Note that since every 2*L duration contributes only one point, if m, the step size, is bigger than one, then the number of resultant points actually shrinks by m times in the time domain. The result can be recovered back by the relationship:
S[j]=S[k] for (j=k; j≦k+m-1; j++)
Preferably, m is exactly one and no correction is needed.
In the seventh block, k is compared with M. If k is less than or equal to M, the whole process is repeated for the second duration for k=k+1 and i=i+1 until k>M.
A two-dimensional representation of local frequency change amplitude in the frequency range of interest associated with the cardiac events in question, more approximately 30-55 Hz and, more specifically, in this example, 31.25-46.875 Hz, which is derived from the Spectrogram distribution by integrating the Spectrogram distribution over that frequency range of interest or, more specifically by combining the discrete Spectrogram function contributions from the frequencies in that range of interest is depicted in FIG. 7. Each block of the Spectrogram derived data extending over the time interval from Ri -0.2(Ri+i -R1) to Ri+1 -0.2(Ri+1 -Ri) is analyzed after it is generated to identify at least one wave in the time-derivative impedance signal data dZ/dt(n) associated with the heartbeat Ri.
Preferably, the time of occurrence of the C wave of the heartbeat in question is first identified directly from the processed dZ/dt signal. The C wave has the greatest amplitude in the portion of the processed dz/dt signal associated with the heartbeat Ri (see FIG. 2B). The Spectrogram derived data preferably is analyzed over the time interval in question spanning the single heartbeat Ri to identify the maximum Spectrogram derived data amplitude value, MAX, occurring over that interval. Proceeding backward in time through the Spectrogram derived data from the time of occurrence of the C wave determined from the dZ/dt signal, the first identified Spectrogram derived data value exceeding MAX/2 encountered occurs at the occurrence of the B wave or event and is referred to as MAX1 in the program. The C wave is referred to as MAX2 in the program. The time of occurrence of the X wave is preferably determined from the interval between the determined B and C events. Preferably, the X wave is identified as the first peak in the Spectrogram having a magnitude≧MAX/3 and occurring at least (B-C) seconds after the time of occurrence of the C wave peak, where B and C represent the times of occurrence of each of those two determined wave peaks. It is referred to as MAX3.
Preferably, the system utilizes a VET=B-X, where B and X represent the time of occurrences of those events as determined in the above described manner from the Spectrogram distribution of the dZ/dt signal. Also in the preferred embodiment, dZ/dtmin is determined to be the absolute magnitude of the original dZ/dt signal between the times of the B and C events as derived from the dZ/dt and Spectrogram data signals. VET and dZ/dtmin can thereafter be plugged into either the Kubicek or Sramek equations.
After stroke volume (SV) is determined, for example by either the Kubicek or Sramek equations. cardiac output (CO) is determined by multiplying the stroke volume by the heart rate in beats per minute, which is preferably determined as described above from the EKG signal.
In comparison parallel tests on patients determining the cardiac output values obtained by the preferred system, utilizing the Kubicek model with a body weight correction factor, and the invasive Swan-Ganz thermodilution method, correlation coefficients of about 0.8 or more were found between the cardiac outputs determined by the two methods.
Preferably, the basic Kubicek equation was modified to take into account over/underweight of the various patients involved by multiplying the basic Kubicek equation by a size factor F determined empirically based upon ideal male and female weights.
Ideal female weight (IFW) in pounds was given by the relation:
IFW=0.534*(height in inches)-27.36
Ideal male weight (IMW) in pounds was given by the relation:
IMW=0.534*(height in inches)-17.36
Deviation of a patient's weight from the ideal weight is given by the relation:
deviation=(actual weight-ideal weight)/ideal weight The stroke volume calculated by the Kubicek model was adjusted up or down by a weight factor F, which had the following values for the following indicated deviation:
If deviation<-0.50, F=0.10*Deviation+0.90
If -0.5<or equal Deviation<0.0, F=0.20* Deviation+1.00;
If 0.0<or equal Deviation<0.6, F=0.43* Deviation+1.00; and
If 0.6<or equal Deviation, F=0.60* Deviation+0.95.
Preferably, blood resistivity R in the Kubicek model was set equal to a constant 135 ohm-cm for male and 112 ohm-cm for female patients. It was found that when actual value of blood resistivity was calculated by the empirical equation:
where HEM is the measured hematocrit, that significant underestimation resulted.
Alternatively, in place of either the Kubicek or Sramek model, the processor 30 of the system 20 can be configured to implement other models including the following relation of stroke volume:
SV=(L2 F[weight]/Cc) (Zo /Z2 Z3)
where Cc is the mean thoracic conductivity, Z2 is the impedance to ground (absolute impedance for absolute ground) sensed by one pair of voltage sensing electrodes, for example 22, and Z3 is the absolute impedance to ground sensed by the remaining pair of voltage sensing electrode, for example 23, and Zo is the mean thoracic impedance previously identified. Again, it may be necessary to provide a scale factor, either F(weight) previously identified or another empirically determined factor, to adjust the relation for underweight and overweight individuals. The conductivity Cc is the inverse of the resistivity.
The above proposed relation between stroke volume and impedance is derived from taking into account the concentric positioning of the generally circular heart within an elliptical pair of lungs within an elliptical thoracic wall in the region on which the voltage sensing electrodes are placed for generation of the impedance signals. The impedance value Zo, Z2 and Z3 would be determined from the impedance signals at the peak of the R wave in the EKG signal, which preferably would be determined as has already been described.
The proposed relation continues to use the mean thoracic impedance Zo, like both the Kubicek and Sramek models, and length L like Kubicek, but avoids the use of dZ/dt employed by both of those models. Consequently, this model has significantly reduced sensitivity to noise, motion artifact, respiratory cycles and ventilation effects. Again, it further can be adjusted to compensate for underweight and overweight individuals.
While the preferred system identifies heartbeats from adaptively processed EKG signals, standard, threshold determined EKG heartbeat and other devices and methods for determining heartbeats can be used in impedance signal processing. While one preferred time-frequency distribution and one preferred window function were discussed in detail, other time-frequency distributions and/or window functions might be employed in the present method and device, particularly when greater processing capability becomes available at comparable costs. Furthermore, at least some aspects of the invention lend themselves to parallel processing to improve upon the speed at which data processing and characteristic determination can be accomplished. The various aspects of the presently preferred embodiment are presently preferred based in large part upon data processing costs and the capability of current equipment. The adaptive identification of heartbeats and consequent heart rate determination by the present invention have their own independent value apart from their use in processing cardiac impedance signals. Other biological signals such as EEG, EMG, pulmonary pressure wave, etc. should also be adaptable to either the adaptive signal processing performed on the EKG signal or the time-frequency distribution processing performed on the impedance signal.
While one preferred embodiment of the system and methods of the present invention have been disclosed and at least one additional alternate suggested, one of ordinary skill in the art will recognize that changes may be made to the above-described embodiments of the invention without departing from the broad inventive concepts thereof. It is understood, therefore, that this invention is not limited to the particular embodiment disclosed but is intended to cover any modifications which are within the scope and spirit of the invention as defined by the appended claims.
|Cited Patent||Filing date||Publication date||Applicant||Title|
|US3149627 *||Apr 25, 1962||Sep 22, 1964||Bagno Samuel M||Plethysmograph|
|US3340867 *||Aug 19, 1964||Sep 12, 1967||Univ Minnesota||Impedance plethysmograph|
|US3452743 *||Mar 1, 1965||Jul 1, 1969||Gen Electric||Body impedance bridge|
|US3570474 *||Mar 27, 1968||Mar 16, 1971||Jaernhs Elektriska Ab||Apparatus for quantitative indicating of small and rapid volume changes in a part of an extremity|
|US3730171 *||Sep 23, 1971||May 1, 1973||Namon R||Impedance related blood flow measuring device|
|US3742936 *||Oct 21, 1970||Jul 3, 1973||Ieram Sarl||Rheoplethysmographic device and method of operation|
|US3835839 *||Dec 8, 1972||Sep 17, 1974||Systron Donner Corp||Impedance plethysmograph and flow rate computer adjunct and method for use therewith|
|US3835840 *||Sep 27, 1973||Sep 17, 1974||Hope City||Impedance plethysmography method and apparatus|
|US3847142 *||Aug 30, 1973||Nov 12, 1974||J Hartwell||Blood flow measuring apparatus and method|
|US3871359 *||Jun 25, 1973||Mar 18, 1975||Interscience Technology Corp||Impedance measuring system|
|US3874368 *||Apr 19, 1973||Apr 1, 1975||Manfred Asrican||Impedance plethysmograph having blocking system|
|US3882851 *||Aug 15, 1973||May 13, 1975||Systron Donner Corp||Impedance plethysmograph|
|US3976052 *||Jan 10, 1975||Aug 24, 1976||Hewlett-Packard Gmbh||Respiration monitor|
|US3994284 *||Dec 31, 1975||Nov 30, 1976||Systron Donner Corporation||Flow rate computer adjunct for use with an impedance plethysmograph and method|
|US3996925 *||May 5, 1975||Dec 14, 1976||Ljubomir Djordjevich||System for determining characteristics of blood flow|
|US4204545 *||Apr 21, 1978||May 27, 1980||Asahi Medical Co., Ltd.||Limb blood flowmeter|
|US4205688 *||May 23, 1977||Jun 3, 1980||Doll Research, Inc.||Method and apparatus for developing and measuring pulsed blood flow|
|US4258720 *||Feb 9, 1979||Mar 31, 1981||Medasonics||Strain gauge plethysmograph|
|US4305400 *||Oct 15, 1979||Dec 15, 1981||Squibb Vitatek Inc.||Respiration monitoring method and apparatus including cardio-vascular artifact detection|
|US4422458 *||Dec 1, 1981||Dec 27, 1983||Montefiore Hospital And Medical Center, Inc.||Method and apparatus for detecting respiratory distress|
|US4432374 *||Nov 25, 1981||Feb 21, 1984||Hiroshi Osanai||Plethysmographic acceleration pulse wave meter|
|US4437469 *||Sep 29, 1980||Mar 20, 1984||Rush-Presbyterian-St. Luke's Medical Center||System for determining characteristics of blood flow|
|US4450527 *||Jun 29, 1982||May 22, 1984||Bomed Medical Mfg. Ltd.||Noninvasive continuous cardiac output monitor|
|US4548211 *||Jan 12, 1984||Oct 22, 1985||Marks Lloyd A||Computer assisted admittance plethysmograph|
|US4562843 *||Jul 6, 1981||Jan 7, 1986||Ljubomir Djordjevich||System for determining characteristics of blood flow|
|US4641260 *||Dec 5, 1983||Feb 3, 1987||Matsushita Electric Industrial Co., Ltd.||Digital signal processing apparatus for a blood flowmeter using ultrasound Doppler effect|
|US4649932 *||Apr 13, 1984||Mar 17, 1987||National Research Development Corp.||Method and apparatus for deriving currents and potentials representative of the impedances of zones of a body|
|US4676253 *||Jul 18, 1985||Jun 30, 1987||Doll Medical Research, Inc.||Method and apparatus for noninvasive determination of cardiac output|
|US4733670 *||Jul 28, 1986||Mar 29, 1988||The Kendall Company||Device for cardiac output monitoring|
|US4757824 *||Aug 15, 1986||Jul 19, 1988||Kontron Holding A.G.||Method and apparatus for monitoring respiration|
|US4805621 *||Jun 15, 1987||Feb 21, 1989||Siemens Aktiengesellschaft||Apparatus for measuring impedance of body tissue|
|US4807638 *||Oct 21, 1987||Feb 28, 1989||Bomed Medical Manufacturing, Ltd.||Noninvasive continuous mean arterial blood prssure monitor|
|US4823797 *||Jun 15, 1987||Apr 25, 1989||Siemens Aktiengesellschaft||Apparatus and method for impedance measurement of body tissues|
|US4836214 *||Dec 1, 1986||Jun 6, 1989||Bomed Medical Manufacturing, Ltd.||Esophageal electrode array for electrical bioimpedance measurement|
|US4870578 *||Aug 17, 1988||Sep 26, 1989||Bomed Medical Manufacturing, Ltd.||Diastolic clamp for bioimpedance measuring device|
|US4905705 *||Mar 3, 1989||Mar 6, 1990||Research Triangle Institute||Impedance cardiometer|
|US4979110 *||Sep 22, 1988||Dec 18, 1990||Massachusetts Institute Of Technology||Characterizing the statistical properties of a biological signal|
|US5025784 *||Sep 6, 1988||Jun 25, 1991||Harbin Polytechnic University||Apparatus and method for detecting and processing impedance rheogram|
|US5046504 *||Apr 6, 1990||Sep 10, 1991||Corazonix Corporation||Method and apparatus for analyzing and interpreting electrocardiograms using spectro-temporal mapping|
|US5101828 *||Apr 11, 1991||Apr 7, 1992||Rutgers, The State University Of Nj||Methods and apparatus for nonivasive monitoring of dynamic cardiac performance|
|US5103828 *||Nov 28, 1990||Apr 14, 1992||Bomed Medical Manufacturing, Ltd.||System for therapeutic management of hemodynamic state of patient|
|US5109862 *||Feb 21, 1991||May 5, 1992||Del Mar Avionics||Method and apparatus for spectral analysis of electrocardiographic signals|
|US5178151 *||Sep 16, 1991||Jan 12, 1993||Sackner Marvin A||System for non-invasive detection of changes of cardiac volumes and aortic pulses|
|US5178154 *||Sep 18, 1990||Jan 12, 1993||Sorba Medical Systems, Inc.||Impedance cardiograph and method of operation utilizing peak aligned ensemble averaging|
|USRE30101 *||Jun 23, 1975||Sep 25, 1979||Regents Of The University Of Minnesota||Impedance plethysmograph|
|1||Geddes, "The Measurement of Cardiac Output and Blood Flow", from Cardiovascular Devices and Their Applications, pp. 100-135.|
|2||*||Geddes, The Measurement of Cardiac Output and Blood Flow , from Cardiovascular Devices and Their Applications, pp. 100 135.|
|3||Gollan et al, "Continuous Electrode Monitoring etc.", British Heart Journal, 1978, pp. 1390-1396.|
|4||*||Gollan et al, Continuous Electrode Monitoring etc. , British Heart Journal, 1978, pp. 1390 1396.|
|5||Kizakevich et al, "An Automated System for Systolic Time Interval Analysis", Proc. Dig. Eq. Comp. Users Soc., 1976, pp. 795-798.|
|6||Kizakevich et al, "Continuous Noninvasive Cardiac Monitoring etc.", Proc. Symp. on Computers In App. Med. Care, 1977 pp. 325-336.|
|7||*||Kizakevich et al, An Automated System for Systolic Time Interval Analysis , Proc. Dig. Eq. Comp. Users Soc., 1976, pp. 795 798.|
|8||*||Kizakevich et al, Continuous Noninvasive Cardiac Monitoring etc. , Proc. Symp. on Computers In App. Med. Care, 1977 pp. 325 336.|
|9||Kubicek et al, "Development and Evaluation of an Impedance Cardiac Output System", Aerospace Medicine, 1966, pp. 1208-1212.|
|10||Kubicek et al, "The Minnesota Impedance Cardiogaph-Theory and Applications", Biomed. Eng. 1974, pp. 410-416.|
|11||*||Kubicek et al, Development and Evaluation of an Impedance Cardiac Output System , Aerospace Medicine, 1966, pp. 1208 1212.|
|12||*||Kubicek et al, The Minnesota Impedance Cardiogaph Theory and Applications , Biomed. Eng. 1974, pp. 410 416.|
|13||Miyamoto et al, "Automatic Determination of Cardiac Output, etc." Biotelemetry Patient Monitoring, 1981, pp. 189-203.|
|14||Miyamoto et al, "Automatic Determination of Cardiac Output, etc." Proc. 5th Int'l Conf. on Electr. Bio-Impedance, 1981, pp. 45-48.|
|15||Miyamoto et al, "Continuous Determination of Cardiac Output, etc.", Med. & Biol. Eng. & Computing, 1981, pp. 638-644.|
|16||*||Miyamoto et al, Automatic Determination of Cardiac Output, etc. Biotelemetry Patient Monitoring, 1981, pp. 189 203.|
|17||*||Miyamoto et al, Automatic Determination of Cardiac Output, etc. Proc. 5th Int l Conf. on Electr. Bio Impedance, 1981, pp. 45 48.|
|18||*||Miyamoto et al, Continuous Determination of Cardiac Output, etc. , Med. & Biol. Eng. & Computing, 1981, pp. 638 644.|
|19||Photocopy of Article "Digital Enhancement of the Peripheral Admittance Plethysmogram", L. A. Marks, IEEE Transactions on Biomedical Engineering, vol. BME-34, No. 3, Mar. 1987 (7 pages).|
|20||Photocopy of article "Mapping the cardiogenic impedance signal on the thoracic surface", R. P. Patterson et al., Medical & Biological Engineering & Computing, May 1990, 5 pages (beginning with p. 212).|
|21||*||Photocopy of Article Digital Enhancement of the Peripheral Admittance Plethysmogram , L. A. Marks, IEEE Transactions on Biomedical Engineering, vol. BME 34, No. 3, Mar. 1987 (7 pages).|
|22||*||Photocopy of article Mapping the cardiogenic impedance signal on the thoracic surface , R. P. Patterson et al., Medical & Biological Engineering & Computing, May 1990, 5 pages (beginning with p. 212).|
|23||Photocopy of Instruction Manual, "Minnesota Impedance Cardiograph-A Noninvasive System to Monitor Cardiovascular Parameters", 4 sheets, Surcom, Inc., 4180 Edmund Boulevard, Minneapolis, Minn. 55406.|
|24||*||Photocopy of Instruction Manual, Minnesota Impedance Cardiograph A Noninvasive System to Monitor Cardiovascular Parameters , 4 sheets, Surcom, Inc., 4180 Edmund Boulevard, Minneapolis, Minn. 55406.|
|25||Sramek et al., "Stroke Volume Equation etc.", Proc. 6th Int'l Conf. on Electr. Bioimpedance, 1983, pp. 1-2.|
|26||*||Sramek et al., Stroke Volume Equation etc. , Proc. 6th Int l Conf. on Electr. Bioimpedance, 1983, pp. 1 2.|
|27||Sramek, "Noninvasive Technique for Meas. of Cardiac Output, etc." Proc. 5th Int'l Conf. on Electr. Bioimpedance, 1981, pp. 39-42.|
|28||*||Sramek, Noninvasive Technique for Meas. of Cardiac Output, etc. Proc. 5th Int l Conf. on Electr. Bioimpedance, 1981, pp. 39 42.|
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US5421345 *||Dec 10, 1993||Jun 6, 1995||Siemens Aktiengesellschaft||Method and apparatus for non-invasive identification of the endocorporeal spatial distribution of the electrical impedance in a subject|
|US5423326 *||May 14, 1993||Jun 13, 1995||Drexel University||Apparatus and method for measuring cardiac output|
|US5503157 *||Mar 17, 1995||Apr 2, 1996||Sramek; Bohumir||System for detection of electrical bioimpedance signals|
|US5505209 *||Jul 7, 1994||Apr 9, 1996||Reining International, Ltd.||Impedance cardiograph apparatus and method|
|US5685316 *||Apr 8, 1996||Nov 11, 1997||Rheo-Graphic Pte Ltd.||Non-invasive monitoring of hemodynamic parameters using impedance cardiography|
|US5735284 *||Jun 13, 1995||Apr 7, 1998||N.I. Medical Ltd.||Method and system for non-invasive determination of the main cardiorespiratory parameters of the human body|
|US5782884 *||Nov 5, 1996||Jul 21, 1998||Sulzer Intermedics Inc.||Rate responsive cardiac pacemaker with peak impedance detection for rate control|
|US5879308 *||Jan 23, 1998||Mar 9, 1999||Instrumentarium Oy||Procedure for measuring a patient's impedance|
|US5913826 *||Jun 12, 1996||Jun 22, 1999||K-One Technologies||Wideband external pulse cardiac monitor|
|US6048319 *||Oct 1, 1998||Apr 11, 2000||Integrated Medical Systems, Inc.||Non-invasive acoustic screening device for coronary stenosis|
|US6058325 *||Aug 19, 1996||May 2, 2000||Cardiotronics||Method and apparatus for high current electrode, transthoracic and transmyocardial impedance estimation|
|US6161038 *||Apr 7, 1997||Dec 12, 2000||Rheo-Graphic Pte Ltd.||Non-invasive monitoring of hemodynamic parameters using impedance cardiography|
|US6253103||Oct 8, 1999||Jun 26, 2001||Kimberly-Clark Worldwide, Inc.||Method and apparatus for high current electrode, transthoracic and transmyocardial impedance estimation|
|US6322518||Feb 9, 1998||Nov 27, 2001||Heska Corporation||Method and apparatus for measuring cardiac output|
|US6438400||Feb 22, 2000||Aug 20, 2002||Heska Corporation||Electrode for evaluating cardiac functions via esophagus|
|US6478744||Dec 18, 2000||Nov 12, 2002||Sonomedica, Llc||Method of using an acoustic coupling for determining a physiologic signal|
|US6511438||Apr 3, 2001||Jan 28, 2003||Osypka Medical Gmbh||Apparatus and method for determining an approximation of the stroke volume and the cardiac output of the heart|
|US6561986||Jan 17, 2001||May 13, 2003||Cardiodynamics International Corporation||Method and apparatus for hemodynamic assessment including fiducial point detection|
|US6625484||Mar 12, 2001||Sep 23, 2003||BIOTRONIK MESS- UND THERAPIEGERäTE GMBH & CO. INGENIEURBüRO BERLIN||Signal evaluation method for detecting QRS complexes in electrocardiogram signals|
|US6643546||Feb 13, 2002||Nov 4, 2003||Quetzal Biomedical, Inc.||Multi-electrode apparatus and method for treatment of congestive heart failure|
|US6829501 *||Dec 20, 2001||Dec 7, 2004||Ge Medical Systems Information Technologies, Inc.||Patient monitor and method with non-invasive cardiac output monitoring|
|US6845263 *||Jul 2, 2001||Jan 18, 2005||Colin Medical Technology Corporation||Heart-sound detecting apparatus and pulse-wave-propagation-velocity-relating-information obtaining system using the heart-sound detecting apparatus|
|US6896660||Jun 19, 2002||May 24, 2005||University Of Southern California||Therapeutic decisions systems and method using stochastic techniques|
|US6907288||Jun 12, 2001||Jun 14, 2005||Cardiac Pacemakers, Inc.||Cardiac rhythm management system adjusting rate response factor for treating hypotension|
|US6912420||Apr 10, 2001||Jun 28, 2005||Cardiac Pacemakers, Inc.||Cardiac rhythm management system for hypotension|
|US7015701 *||Mar 4, 2003||Mar 21, 2006||Gerald Wiegand||Highly time resolved impedance spectroscopy|
|US7043293||Dec 24, 2002||May 9, 2006||Cardiodynamics International Corporation||Method and apparatus for waveform assessment|
|US7054679 *||Oct 31, 2001||May 30, 2006||Robert Hirsh||Non-invasive method and device to monitor cardiac parameters|
|US7177681||Apr 11, 2003||Feb 13, 2007||Cardiac Pacemakers, Inc.||Cardiac rhythm management system for edema|
|US7186219||Oct 10, 2002||Mar 6, 2007||Osypka Medical Gmbh||Calibration of a doppler velocimeter for stroke volume determination|
|US7191000||Jul 31, 2001||Mar 13, 2007||Cardiac Pacemakers, Inc.||Cardiac rhythm management system for edema|
|US7200441||Jun 21, 2004||Apr 3, 2007||Cardiac Pacemakers, Inc.||Implantable cardiac stimulator with electrode-tissue interface characterization|
|US7261697||Jun 16, 2004||Aug 28, 2007||Bernstein Donald P||Apparatus for determination of stroke volume using the brachial artery|
|US7340296||May 18, 2005||Mar 4, 2008||Cardiac Pacemakers, Inc.||Detection of pleural effusion using transthoracic impedance|
|US7351203||Jan 18, 2005||Apr 1, 2008||University Of Southern California||Therapeutic decisions systems and method using stochastic techniques|
|US7387610||Aug 19, 2004||Jun 17, 2008||Cardiac Pacemakers, Inc.||Thoracic impedance detection with blood resistivity compensation|
|US7395114 *||Aug 20, 2004||Jul 1, 2008||Biotronik Gmbh & Co.||Intracardial impedance measuring arrangement|
|US7474918 *||Mar 24, 2005||Jan 6, 2009||Noninvasive Medical Technologies, Inc.||Thoracic impedance monitor and electrode array and method of use|
|US7570989 *||Nov 22, 2004||Aug 4, 2009||Cardiodynamics International Corporation||Method and apparatus for signal assessment including event rejection|
|US7603170||Oct 13, 2009||Cardiac Pacemakers, Inc.||Calibration of impedance monitoring of respiratory volumes using thoracic D.C. impedance|
|US7657306||Feb 2, 2010||Robert Hirsh||Non-invasive method and device to monitor cardiac parameters|
|US7672718||Jun 16, 2008||Mar 2, 2010||Cardiac Pacemakers, Inc.||Thoracic impedance detection with blood resistivity compensation|
|US7740590||Jun 22, 2009||Jun 22, 2010||Cordeus, Inc.||Apparatus and method for determination of stroke volume using the brachial artery|
|US7761141||Mar 22, 2007||Jul 20, 2010||Robert Hirsh||Non-invasive method and device to monitor cardiac parameters without use of electrical-mechanical interval|
|US7806830||Jun 21, 2005||Oct 5, 2010||Cordeus, Inc.||Apparatus and method for determination of stroke volume using the brachial artery|
|US7822470||Oct 9, 2002||Oct 26, 2010||Osypka Medical Gmbh||Method for determining the left-ventricular ejection time TLVE of a heart of a subject|
|US7881781||Feb 1, 2011||Cardiac Pacemakers, Inc.||Thoracic impedance detection with blood resistivity compensation|
|US7904141||Mar 22, 2006||Mar 8, 2011||Osypka Medical Gmbh||System and apparatus for determining the left-ventricular ejection time TLVE of a heart of a subject|
|US7907997||Mar 15, 2011||Cardiac Pacemakers, Inc.||Enhancements to the detection of pulmonary edema when using transthoracic impedance|
|US7979115||Jul 12, 2011||Cardiac Pacemakers, Inc.||Detection of pleural effusion using transthoracic impedance|
|US7983749 *||Jul 19, 2011||Cardiac Pacemakers, Inc.||Cardiac rhythm management system with time-dependent frequency response|
|US8050764||Oct 29, 2003||Nov 1, 2011||Cardiac Pacemakers, Inc.||Cross-checking of transthoracic impedance and acceleration signals|
|US8068906||Nov 29, 2011||Aorora Technologies Pty Ltd||Cardiac monitoring system|
|US8099250||Jul 27, 2006||Jan 17, 2012||Impedimed Limited||Impedance parameter values|
|US8111152||Mar 23, 2009||Feb 7, 2012||Noninvasive Medical Technologies, Inc.||Relative positioning system and method|
|US8134460||Mar 23, 2009||Mar 13, 2012||Noninvasive Medical Technologies, Inc.||Relative positioning system method|
|US8157848||Jan 21, 2009||Apr 17, 2012||Siemens Medical Solutions Usa, Inc.||System for characterizing patient tissue impedance for monitoring and treatment|
|US8200326||Jun 12, 2012||Cardiac Pacemakers, Inc.||Calibration of impedance monitoring of respiratory volumes using thoracic D.C. impedance|
|US8216136||Jul 10, 2012||Nellcor Puritan Bennett Llc||Systems and methods for monitoring heart rate and blood pressure correlation|
|US8219198 *||Aug 16, 2010||Jul 10, 2012||Medtronic, Inc.||Method and device for using impedance measurements based on electrical energy of the heart|
|US8233974||Jan 29, 2004||Jul 31, 2012||Impedimed Limited||Method and device for measuring tissue oedema|
|US8271079||Feb 8, 2010||Sep 18, 2012||Biospace Co., Ltd.||Method of analyzing body composition with measurement of voltage signals at multiple positions of body|
|US8290730||Oct 16, 2012||Nellcor Puritan Bennett Ireland||Systems and methods for assessing measurements in physiological monitoring devices|
|US8306621||Nov 6, 2012||Cardiac Pacemakers, Inc.||Cardiac cycle synchronized sampling of impedance signal|
|US8332026||Feb 8, 2010||Dec 11, 2012||Biospace Co., Ltd.||Method of analyzing body composition with measurement of voltage signals at multiple positions of body|
|US8386028||Feb 26, 2013||Biospace Co., Ltd.||Method of analyzing body composition with measurement of voltage signals at multiple positions of body|
|US8388545||Jan 18, 2006||Mar 5, 2013||Cheetah Medical, Inc.||System, method and apparatus for measuring blood flow and blood volume|
|US8398556||Sep 30, 2008||Mar 19, 2013||Covidien Lp||Systems and methods for non-invasive continuous blood pressure determination|
|US8414498||May 10, 2004||Apr 9, 2013||Cheetah Medical, Inc.||System, method and apparatus for measuring blood flow and blood volume|
|US8423142||Oct 31, 2011||Apr 16, 2013||Cardiac Pacemakers, Inc.||Cross-checking of transthoracic impedance and acceleration signals|
|US8442633||Oct 30, 2012||May 14, 2013||Cardiac Pacemakers, Inc.||Cardiac cycle synchronized sampling of impedance signal|
|US8463347||Sep 30, 2009||Jun 11, 2013||Nellcor Puritan Bennett Ireland||Systems and methods for normalizing a plethysmograph signal for improved feature analysis|
|US8483818||Feb 23, 2011||Jul 9, 2013||Cardiac Pacemakers, Inc.||Enhancements to the detection of pulmonary edema when using transthoracic impedance|
|US8487686||Mar 31, 2008||Jul 16, 2013||Impedimed Limited||Active guarding for reduction of resistive and capacitive signal loading with adjustable control of compensation level|
|US8506498||Oct 9, 2008||Aug 13, 2013||Nellcor Puritan Bennett Ireland||Systems and methods using induced perturbation to determine physiological parameters|
|US8509886||Nov 28, 2011||Aug 13, 2013||Aorora Technologies Pty Ltd||Cardiac monitoring system|
|US8523777||Apr 15, 2008||Sep 3, 2013||Cheetah Medical, Inc.||Method, apparatus and system for predicting electromechanical dissociation|
|US8532751||Sep 30, 2008||Sep 10, 2013||Covidien Lp||Laser self-mixing sensors for biological sensing|
|US8548580||Jun 30, 2006||Oct 1, 2013||Impedimed Limited||Monitoring system|
|US8560245||Nov 21, 2008||Oct 15, 2013||Nellcor Puritan Bennett Ireland||Continuous non-invasive blood pressure measurement apparatus and methods providing automatic recalibration|
|US8562538||Mar 7, 2011||Oct 22, 2013||Osypka Medical Gmbh||System for determining the left-ventricular ejection time TLVE of a heart of a subject|
|US8594781||Jan 14, 2008||Nov 26, 2013||Impedimed Limited||Monitoring system|
|US8628477||Jul 31, 2009||Jan 14, 2014||Nellcor Puritan Bennett Ireland||Systems and methods for non-invasive determination of blood pressure|
|US8660799||Oct 3, 2008||Feb 25, 2014||Nellcor Puritan Bennett Ireland||Processing and detecting baseline changes in signals|
|US8682424||Jul 29, 2004||Mar 25, 2014||Dst Delta Segments Technology, Inc.||Noninvasive multi-channel monitoring of hemodynamic parameters|
|US8688214||May 10, 2013||Apr 1, 2014||Cardiac Pacemakers. Inc.||Cardiac cycle synchronized sampling of impedance signal|
|US8692717||Mar 23, 2009||Apr 8, 2014||Noninvasive Medical Technologies, Inc.||Antenna for thoracic radio interrogation|
|US8700121||Dec 14, 2012||Apr 15, 2014||Intersection Medical, Inc.||Devices for determining the relative spatial change in subsurface resistivities across frequencies in tissue|
|US8761870||May 25, 2007||Jun 24, 2014||Impedimed Limited||Impedance measurements|
|US8764667||Mar 6, 2008||Jul 1, 2014||Cheetah Medical, Inc.||Method and system for monitoring sleep|
|US8781551||Jun 30, 2006||Jul 15, 2014||Impedimed Limited||Apparatus for connecting impedance measurement apparatus to an electrode|
|US8781587 *||Jan 19, 2013||Jul 15, 2014||Eckhard Alt||Detecting and treatment of sleep apnea|
|US8790267||Aug 11, 2008||Jul 29, 2014||Cheetah Medical, Inc.||Dynamically variable filter|
|US8798715||Jan 10, 2010||Aug 5, 2014||Aspect Imaging Ltd||Method for non-invasive measurement of cardiac output|
|US8808191||Jul 27, 2009||Aug 19, 2014||Robert Hirsh||Non-invasive method and device to monitor cardiac parameters|
|US8825428||Nov 30, 2010||Sep 2, 2014||Neilcor Puritan Bennett Ireland||Methods and systems for recalibrating a blood pressure monitor with memory|
|US8836345||Oct 15, 2008||Sep 16, 2014||Impedimed Limited||Impedance determination|
|US8843194||Dec 8, 2009||Sep 23, 2014||Robert Hirsh||Non-invasive method and device to monitor cardiac parameters|
|US8876725||Feb 21, 2008||Nov 4, 2014||Cheetah Medical, Inc.||Method and system for estimating exercise capacity|
|US8880171||Feb 13, 2014||Nov 4, 2014||Cardiac Pacemakers, Inc.||Cardiac cycle synchronized sampling of impedance signal|
|US8898037||Apr 28, 2010||Nov 25, 2014||Nellcor Puritan Bennett Ireland||Systems and methods for signal monitoring using Lissajous figures|
|US8932219||Jun 20, 2012||Jan 13, 2015||Nellcor Puritan Bennett Ireland||Systems and methods for monitoring heart rate and blood pressure correlation|
|US9050016||Feb 9, 2010||Jun 9, 2015||Siemens Medical Solutions Usa, Inc.||System for heart performance characterization and abnormality detection|
|US9060695||Nov 30, 2011||Jun 23, 2015||Covidien Lp||Systems and methods for determining differential pulse transit time from the phase difference of two analog plethysmographs|
|US9066660||Sep 29, 2009||Jun 30, 2015||Nellcor Puritan Bennett Ireland||Systems and methods for high-pass filtering a photoplethysmograph signal|
|US9089275||May 11, 2005||Jul 28, 2015||Cardiac Pacemakers, Inc.||Sensitivity and specificity of pulmonary edema detection when using transthoracic impedance|
|US9095271||Aug 13, 2007||Aug 4, 2015||Cheetah Medical, Inc.||Dynamically variable filter|
|US9113805||Mar 4, 2013||Aug 25, 2015||Mortara Instrument, Inc.||Impedance measurement system|
|US9149225||Dec 14, 2012||Oct 6, 2015||Intesection Medical, Inc.||Methods for determining the relative spatial change in subsurface resistivities across frequencies in tissue|
|US9149235||Dec 20, 2013||Oct 6, 2015||Impedimed Limited||Oedema detection|
|US9198582||Jun 30, 2009||Dec 1, 2015||Nellcor Puritan Bennett Ireland||Determining a characteristic physiological parameter|
|US9220440||Sep 21, 2009||Dec 29, 2015||Nellcor Puritan Bennett Ireland||Determining a characteristic respiration rate|
|US9259160||Dec 1, 2010||Feb 16, 2016||Nellcor Puritan Bennett Ireland||Systems and methods for determining when to measure a physiological parameter|
|US9265950||May 12, 2008||Feb 23, 2016||Cardiac Pacemakers, Inc.||Contractility modulation based on impedance signal|
|US9289136||Aug 26, 2014||Mar 22, 2016||Nellcor Puritan Bennett Ireland||Methods and systems for recalibrating a blood pressure monitor with memory|
|US9289167||Dec 5, 2012||Mar 22, 2016||Masimo Corporation||Signal processing apparatus and method|
|US9297874||Mar 12, 2014||Mar 29, 2016||Aspect Imaging Ltd.||Apparatus and method for non-invasive measurement of cardiac output|
|US9301697||Sep 30, 2008||Apr 5, 2016||Nellcor Puritan Bennett Ireland||Systems and methods for recalibrating a non-invasive blood pressure monitor|
|US9314168||Sep 30, 2008||Apr 19, 2016||Nellcor Puritan Bennett Ireland||Detecting sleep events using localized blood pressure changes|
|US20020147475 *||Apr 10, 2001||Oct 10, 2002||Avram Scheiner||Cardiac rhythm management system for hypotension|
|US20020147476 *||Jun 12, 2001||Oct 10, 2002||Daum Douglas R.||Cardiac rhythm management system adjusting rate response factor for treating hypotension|
|US20020169484 *||Feb 13, 2002||Nov 14, 2002||Scott Mathis||Multi-electrode apparatus and method for treatment of congestive heart failure|
|US20030009108 *||Jul 2, 2001||Jan 9, 2003||Colin Corporation||Heart-sound detecting apparatus and pulse-wave-propagation-velocity-relating-information obtaining system using the heart-sound detecting apparatus|
|US20030028221 *||Jul 31, 2001||Feb 6, 2003||Qingsheng Zhu||Cardiac rhythm management system for edema|
|US20030060690 *||Jun 19, 2002||Mar 27, 2003||Roger Jelliffe||Therapeutic decisions systems and method using stochastic techniques|
|US20030083582 *||Oct 31, 2001||May 1, 2003||Robert Hirsh||Non-invasive method and device to monitor cardiac parameters|
|US20030163058 *||Oct 9, 2002||Aug 28, 2003||Osypka Markus J.||Method and apparatus for determining the left-ventricular ejection time TLVE of a heart of a subject|
|US20030173976 *||Mar 4, 2003||Sep 18, 2003||Gerald Wiegand||Highly time resolved impedance spectroscopy|
|US20030191503 *||Apr 11, 2003||Oct 9, 2003||Cardiac Pacemakers, Inc.||Cardiac rhythm management system for edema|
|US20040152996 *||May 23, 2002||Aug 5, 2004||Eberhard Gersing||Transformer-isolated alternating current power supply|
|US20040167578 *||Jul 10, 2003||Aug 26, 2004||Warren Jay A.||Cardiac rhythm management system with time-dependent frequency response|
|US20040186392 *||Jan 29, 2004||Sep 23, 2004||The University Of Queensland||Method and device for measuring tissue oedema|
|US20040230245 *||Jun 21, 2004||Nov 18, 2004||Intermedics Inc.||Implantable cardiac stimulator with electrode-tissue interface characterization|
|US20050012414 *||Jul 16, 2004||Jan 20, 2005||Osypka Medical Gmbh||Method and apparatus for isolated transformation of a first voltage into a second voltage for measurement of electrical bioimpedances or bioconductances|
|US20050096704 *||Oct 29, 2003||May 5, 2005||Scott Freeberg||Cross-checking of transthoracic impedence and acceleration signals|
|US20050143633 *||Jan 18, 2005||Jun 30, 2005||University Of Southern California, A Corporation||Therapeutic decisions systems and method using stochastic techniques|
|US20050203427 *||Mar 15, 2004||Sep 15, 2005||Impedance Vascular Imaging Products, Llc||Stenosis detection device|
|US20050203428 *||Mar 15, 2004||Sep 15, 2005||Impedance Vascular Imaging Products, Llc||Device and method for determining coronary blood flow|
|US20050203429 *||Mar 15, 2004||Sep 15, 2005||Impedance Vascular Imaging Products, Llc||Device and method for measuring cardiac function|
|US20050215918 *||Mar 24, 2005||Sep 29, 2005||Frantz Ann K||Thoracic impedance monitor and electrode array and method of use|
|US20050283088 *||Jun 16, 2004||Dec 22, 2005||Bernstein Donald P||Apparatus and method for determination of stroke volume using the brachial artery|
|US20060009710 *||Jun 21, 2005||Jan 12, 2006||Bernstein Donald P||Apparatus and method for determination of stroke volume using the brachial artery|
|US20060041280 *||Aug 19, 2004||Feb 23, 2006||Cardiac Pacemakers, Inc.||Thoracic impedance detection with blood resistivity compensation|
|US20060064026 *||Nov 3, 2005||Mar 23, 2006||Robert Hirsh||Non-invasive method and device to monitor cardiac parameters|
|US20060111641 *||Nov 19, 2004||May 25, 2006||Applied Cardiac Systems, Inc.||System and method for ICG recording and analysis|
|US20060111642 *||Nov 22, 2004||May 25, 2006||Baura Gail D||Method and apparatus for signal assessment including event rejection|
|US20060122523 *||Oct 17, 2003||Jun 8, 2006||Giorgio Bonmassar||Arrangement and method for detecting abnormalities and inconsistencies in a body|
|US20060167363 *||Mar 22, 2006||Jul 27, 2006||Osypka Medical Gmbh||System and apparatus for determining the left-ventricular ejection time TLVE of a heart of a subject|
|US20060200033 *||May 10, 2004||Sep 7, 2006||Cheetah Medical Inc. C/O Pepper Hamiton||System, method and apparatus for measuring blood flow and blood volume|
|US20060241513 *||Apr 26, 2005||Oct 26, 2006||Cardiac Pacemakers, Inc.||Calibration of impedance monitoring of respiratory volumes using thoracic D.C. impedance|
|US20060247543 *||Oct 9, 2003||Nov 2, 2006||Bruce Cornish||High resoution bio-impedance device|
|US20060258952 *||May 11, 2005||Nov 16, 2006||Cardiac Pacemakers, Inc.||Enhancements to the detection of pulmonary edema when using transthoracic impedance|
|US20060271024 *||May 9, 2006||Nov 30, 2006||Michael Gertner||Nasal Cavity Treatment Apparatus|
|US20060293609 *||May 11, 2005||Dec 28, 2006||Cardiac Pacemakers, Inc.||Sensitivity and specificity of pulmonary edema detection when using transthoracic impedance|
|US20070043303 *||Aug 17, 2006||Feb 22, 2007||Osypka Markus J||Method and apparatus for digital demodulation and further processing of signals obtained in the measurement of electrical bioimpedance or bioadmittance in an object|
|US20070129639 *||Jan 10, 2005||Jun 7, 2007||Hongxuan Zhang||Methods and analysis for cardiac ischemia detection|
|US20070142867 *||Feb 16, 2007||Jun 21, 2007||Cardiac Pacemakers, Inc.||Cardiac cycle synchronized sampling of impedance signal|
|US20070191724 *||Mar 22, 2007||Aug 16, 2007||Robert Hirsh||Non-invasive method and device to monitor cardiac parameters without use of electrical-mechanical interval|
|US20080009757 *||Jul 29, 2004||Jan 10, 2008||Alexander Noson Tsoglin||Noninvasive Multi-Channel Monitoring of Hemodynamic Parameters|
|US20080215108 *||May 12, 2008||Sep 4, 2008||Cardiac Pacemakers, Inc.||Cardiac rhythm management system for edema|
|US20080287823 *||Jul 19, 2006||Nov 20, 2008||Scott Matthew Chetham||Index Determination|
|US20090043222 *||Oct 10, 2006||Feb 12, 2009||Scott Chetham||Hydration status monitoring|
|US20090076398 *||Nov 21, 2008||Mar 19, 2009||Nellcor Puritan Bennett Ireland||Continuous Non-Invasive Blood Pressure Measurement Apparatus and Methods Providing Automatic Recalibration|
|US20090143663 *||Jun 30, 2006||Jun 4, 2009||Impedance Cardiology Systems Inc.||Pulmonary Monitoring System|
|US20090198300 *||Jan 21, 2009||Aug 6, 2009||Siemens Medical Solutions Usa, Inc.||System for Characterizing Patient Tissue Impedance for Monitoring and Treatment|
|US20090237234 *||Mar 23, 2009||Sep 24, 2009||Noninvasive Medical Technologies, Inc.||Relative positioning system and method|
|US20090237235 *||Mar 23, 2009||Sep 24, 2009||Noninvasive Medical, Technologies, Inc.||Relative positioning system method|
|US20090240132 *||Mar 23, 2009||Sep 24, 2009||Noninvasive Medical Technologies, Inc.||Antenna for thoracic radio interrogation|
|US20090240133 *||Mar 23, 2009||Sep 24, 2009||Noninvasive Medical Technologies, Inc.||Apparatus and method for non-invasive, in-vivo, thoracic radio interrogation|
|US20090240134 *||Mar 23, 2009||Sep 24, 2009||Andrew Pal||Method of processing thoracic reflected radio interrogation signals|
|US20090287102 *||Feb 13, 2009||Nov 19, 2009||Impedimed Limited||Blood flow assessment of venous insufficiency|
|US20090287105 *||Jul 27, 2009||Nov 19, 2009||Robert Hirsh||Non-invasive method and device to monitor cardiac parameters|
|US20090326353 *||Oct 3, 2008||Dec 31, 2009||Nellcor Puritan Bennett Ireland||Processing and detecting baseline changes in signals|
|US20090326386 *||Sep 30, 2008||Dec 31, 2009||Nellcor Puritan Bennett Ireland||Systems and Methods for Non-Invasive Blood Pressure Monitoring|
|US20090326393 *||Dec 31, 2009||Nellcor Puritan Bennett Ireland||Systems and Methods for Non-Invasive Continuous Blood Pressure Determination|
|US20090326599 *||Dec 31, 2009||John Hatlestad||Calibration of impedance monitoring of respiratory volumes using thoracic d.c. impedance|
|US20100031959 *||Mar 6, 2008||Feb 11, 2010||Cheetah Medical Ltd.||Method and system for monitoring sleep|
|US20100069765 *||Jan 18, 2006||Mar 18, 2010||Cheetah Medical Ltd.||System, Method and Apparatus for Measuring Blood Flow and Blood Volume|
|US20100081940 *||Apr 1, 2010||Nellcor Puritan Bennett Llc||Laser Self-Mixing Sensors for Biological Sensing|
|US20100081943 *||Sep 30, 2008||Apr 1, 2010||Nellcor Puritan Bennett Ireland||Detecting Sleep Events Using Localized Blood Pressure Changes|
|US20100081944 *||Sep 30, 2008||Apr 1, 2010||Nellcor Puritan Bennett Ireland||Systems and Methods for Recalibrating a Non-Invasive Blood Pressure Monitor|
|US20100081945 *||Apr 1, 2010||Nellcor Puritan Bennett Ireland||Systems and Methods for Maintaining Blood Pressure Monitor Calibration|
|US20100087750 *||May 25, 2007||Apr 8, 2010||Mcgree James Matthew||Impedance measurements|
|US20100100003 *||Jan 14, 2008||Apr 22, 2010||Impedimed Limited||Monitoring system|
|US20100109739 *||Mar 31, 2008||May 6, 2010||Impedimed Limited||Active guarding for reduction of resistive and capacitive signal loading with adjustable control of compensation level|
|US20100140153 *||Nov 2, 2009||Jun 10, 2010||Jacob Telepciak||Manifold block for reverse osmosis systems|
|US20100152605 *||Apr 17, 2008||Jun 17, 2010||Impedimed Limited||Monitoring system and probe|
|US20100168530 *||Nov 29, 2007||Jul 1, 2010||Impedimed Limited||Measurement apparatus|
|US20100191127 *||Feb 21, 2008||Jul 29, 2010||Cheetah Mmedical Ltd.||Method and system for estimating exercise capacity|
|US20100204585 *||Feb 9, 2010||Aug 12, 2010||Siemens Medical Solutions Usa, Inc.||System for Heart Performance Characterization and Abnormality Detection|
|US20100217140 *||Apr 15, 2008||Aug 26, 2010||Cheetah Medical Ltd.||Method, apparatus and system for predicting electromechanical dissociation|
|US20100228140 *||Sep 9, 2010||Robert Hirsh||Non-invasive method and device to monitor cardiac parameters|
|US20100249640 *||Sep 30, 2010||Biospace Co., Ltd.|
|US20100249641 *||Sep 30, 2010||Biospace Co., Ltd.|
|US20100249642 *||Feb 8, 2010||Sep 30, 2010||Biospace Co., Ltd.|
|US20100324431 *||Jun 18, 2009||Dec 23, 2010||Nellcor Puritan Bennett Ireland||Determining Disease State Using An Induced Load|
|US20100331724 *||Jun 30, 2009||Dec 30, 2010||Nellcor Puritan Bennett Ireland||Determining a characteristic blood pressure|
|US20100332173 *||Jun 30, 2009||Dec 30, 2010||Nellcor Puritan Bennett Ireland||Systems and methods for assessing measurements in physiological monitoring devices|
|US20110021929 *||Jan 27, 2011||Nellcor Puritan Bennett Ireland||Systems and methods for continuous non-invasive blood pressure monitoring|
|US20110025348 *||Oct 15, 2008||Feb 3, 2011||Impedimed Limited||Impedance determination|
|US20110028854 *||Jul 31, 2009||Feb 3, 2011||Nellcor Puritain Bennett Ireland||Systems and methods for non-invasive determination of blood pressure|
|US20110046505 *||Aug 7, 2008||Feb 24, 2011||Impedimed Limited||Impedance measurement process|
|US20110054343 *||Jun 30, 2006||Mar 3, 2011||Impedimed Limited||Monitoring system|
|US20110071406 *||Sep 21, 2009||Mar 24, 2011||Nellcor Puritan Bennett Ireland||Determining A Characteristic Respiration Rate|
|US20110077486 *||Mar 31, 2011||Nellcor Puritan Bennett Ireland||Systems and methods for normalizing a plethysmograph signal for improved feature analysis|
|US20110077531 *||Sep 29, 2009||Mar 31, 2011||Nellcor Puritan Bennett Ireland||Systems and methods for high-pass filtering a photoplethysmograph signal|
|US20110087119 *||Apr 14, 2011||Maik Gollasch||Method and device for using impedance measurements based on electrical energy of the heart|
|US20110087129 *||Jun 30, 2006||Apr 14, 2011||Impedimed Limited||Monitoring system|
|US20110144526 *||Jun 16, 2011||Stahmann Jeffrey E||Enhancements to the detection of pulmonary edema when using transthoracic impedance|
|US20110172518 *||Jul 14, 2011||Uri Rapoport||Method for Non-Invasive Measurement of Cardiac Output|
|US20110190601 *||Aug 4, 2011||Osypka Markus J||System for Determining the Left-Ventricular Ejection Time TLVE of a Heart of a Subject|
|US20110218419 *||Aug 11, 2008||Sep 8, 2011||Cheetah Medical Ltd.||Dynamically variable filter|
|US20120016254 *||Jan 19, 2012||Tanita Corporation||Respiration characteristic analysis apparatus and respiration characteristic analysis system|
|US20120016255 *||Jan 19, 2012||Tanita Corporation||Respiration characteristic analysis apparatus and respiration characteristic analysis system|
|US20130197376 *||Jan 19, 2013||Aug 1, 2013||Eckhard Alt||Detecting and treatment of sleep apnea|
|US20130204156 *||Jan 25, 2013||Aug 8, 2013||T4 Analytics Llc||Methods and Systems For Assessing Muscle Electrical Activity in Response to Stimulation of a Motor Nerve|
|US20140249435 *||Dec 27, 2011||Sep 4, 2014||Matt Banet||Body-worn system for continuous, noninvasive measurement of cardiac output, stroke volume, cardiac power, and blood pressure|
|US20140249440 *||Dec 27, 2011||Sep 4, 2014||Matt Banet||Body-worn system for continuous, noninvasive measurement of cardiac output, stroke volume, cardiac power, and blood pressure|
|US20140249441 *||Dec 28, 2011||Sep 4, 2014||Matt Banet||Body-worn system for continuous, noninvasive measurement of cardiac output, stroke volume, cardiac power, and blood pressure|
|US20140249443 *||Dec 28, 2011||Sep 4, 2014||Matt Banet|
|CN103083011B *||Jan 22, 2013||Nov 5, 2014||西安交通大学||Electrocardio R wave crest real-time locating method assisted by chest impedance second order difference image|
|EP0782408A2 *||Sep 21, 1995||Jul 9, 1997||Beth Israel Deaconess Medical Center||Fetal data processing system and method|
|EP1132045A2 *||Feb 8, 2001||Sep 12, 2001||BIOTRONIK Mess- und Therapiegeräte GmbH & Co Ingenieurbüro Berlin||Signal evaluation method for detecting QRS complexes in electrocardiogram signals|
|EP1553871A1 *||Oct 9, 2003||Jul 20, 2005||Queensland University of Technology||High resolution bio-impedance device|
|WO1996029003A1 *||Mar 15, 1996||Sep 26, 1996||Bohumir Sramek||System for detection of electrical bioimpedance signals|
|WO2001003580A1 *||Jul 3, 2000||Jan 18, 2001||Instrumentarium Oy||Method for continuous monitoring of cardiac output|
|WO2001062147A1||Feb 14, 2001||Aug 30, 2001||Heska Corporation||Apparatus for evaluating cardiac functions via esophagus|
|WO2002103492A2 *||Jun 19, 2002||Dec 27, 2002||University Of Southern California||Therapeutic decisions systems and method using stochastic techniques|
|WO2002103492A3 *||Jun 19, 2002||Oct 16, 2003||Univ Southern California||Therapeutic decisions systems and method using stochastic techniques|
|WO2004032738A1||Oct 9, 2003||Apr 22, 2004||Queensland University Of Technology||High resolution bio-impedance device|
|WO2004036379A2 *||Oct 17, 2003||Apr 29, 2004||The General Hospital Corporation||Arrangement and method for detecting abnormalities and inconsistencies in a body|
|WO2004036379A3 *||Oct 17, 2003||Jul 1, 2004||John W Belliveau||Arrangement and method for detecting abnormalities and inconsistencies in a body|
|WO2004047629A1 *||Nov 21, 2002||Jun 10, 2004||Cardiobeat.Com||Medical testing system and method|
|WO2005010640A3 *||Jul 29, 2004||Jul 14, 2005||Arkady Hanon Margolin||Noninvasive multi-channel monitoring of hemodynamic parameters|
|U.S. Classification||600/508, 600/526, 600/513, 600/547|
|International Classification||A61B5/05, A61B5/053, A61B5/0295, A61B5/04|
|Cooperative Classification||A61B5/7239, A61B5/04012, A61B5/0535, A61B5/7257|
|European Classification||A61B5/053F, A61B5/04R|
|Apr 16, 1992||AS||Assignment|
Owner name: DREXEL UNIVERSITY, PENNSYLVANIA
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST.;ASSIGNORS:WANG, XIANG;SUN, HUN H.;REEL/FRAME:006083/0978
Effective date: 19920413
|Sep 29, 1997||FPAY||Fee payment|
Year of fee payment: 4
|Sep 28, 2001||FPAY||Fee payment|
Year of fee payment: 8
|Apr 15, 2004||AS||Assignment|
Owner name: NONINVASIVE MEDICAL TECHNOLOGIES, LLC, MICHIGAN
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:DREXEL UNIVERSITY;REEL/FRAME:015215/0449
Effective date: 20040401
|Nov 23, 2005||REMI||Maintenance fee reminder mailed|
|Apr 11, 2006||SULP||Surcharge for late payment|
Year of fee payment: 11
|Apr 11, 2006||FPAY||Fee payment|
Year of fee payment: 12